DFTB+ Lattice Optimization Tool

Overview

The dftb_lattice_opt_tool is a Python script that simplifies the process of batch lattice optimization calculations using the DFTB+ (Density-Functional Tight-Binding) software. It automates the generation of hierarchical directories, assignment of lattice parameters, and creation of DFTB+ input files (.hsd) for multiple structures. Key Features

• Pre-set angular momentum values for common chemical elements
• Ability to add or remove angular momentum for specific elements
• Automatic generation of directories based on the specified gradient and step length
• Assignment of geometry files (.gen) to the appropriate directories
• Generation and assignment of DFTB+ input files (.hsd) with the correct lattice parameters
• Copying of submission scripts (e.g., .sh) to the generated directories

Usage

###Import the necessary modules:

import numpy as np
from ase import io
import ase
import os
from ase.geometry.analysis import Analysis
import shutil
import pandas as pd
import re

###Initialize the lattice_opt class:
lo = lattice_opt()

###Set the required parameters:
lo.save_parameters(
    gradient_number=5,
    step_length=0.01,
    folder_name="lattice_optimization",
    folder_path="/path/to/output/directory",
    output="dftb_in.gen",
    subdir_layernum=2
)

###Prepare the angular momentum dictionary:
angular_m_dic = lo.pre_set_momenta()

###Read the input geometry files:
geo_path_list = lo.read_path("/path/to/input/directory")

###Generate the geometry files with varying lattice parameters:
geo_collect_all = []
for geo_path in geo_path_list:
    geo = io.read(geo_path)
    geo_collect = lo.lattice_sep(geo)
    geo_collect_all.append(geo_collect)

###Create the directories and assign the geometry files:
lo.folders_make(geo_path_list)
lo.assign_geo(geo_collect_all, geo_path_list)

###Generate and assign the DFTB+ input files (.hsd):
hsd_collect = lo.generate_hsd(
    geo_list=geo_collect_all,
    path_hsd_template="/path/to/hsd/template.hsd",
    angular_m_dic=angular_m_dic,
    layer_sep_index=[None, 10, 20]
)
lo.assign_hsd(geo_collect_all, hsd_collect, geo_path_list)

###Assign the submission scripts (e.g., .sh) to the directories:
lo.assign_submit(geo_collect_all, "/path/to/submit/template.sh", geo_path_list)

Customization

• The pre_set_momenta() method provides a dictionary of pre-defined angular momentum values for common elements. You can use the add_momenta() and delete_momenta() methods to customize the angular momentum settings as needed.
• The save_parameters() method allows you to set the specific parameters for your lattice optimization, such as the gradient number, step length, output directory, and more.
• The script assumes the existence of a DFTB+ input file (.hsd) template. You can modify the generate_hsd() method to generate the input files according to your specific requirements.
• The assign_submit() method copies a submission script (e.g., .sh) to the generated directories. You can customize the submission script as needed for your computing environment.

Contributing

Contributions to this project are welcome. If you find any issues or have suggestions for improvements, please feel free to open a new issue or submit a pull request.

License

This project is licensed under the MIT License.